summaryrefslogtreecommitdiff
path: root/gcc/testsuite/gcc.dg/vect/vect-reduc-dot-s8a.c
blob: ac674749b6fe09f3c0adc991c825fb9eacc9eee5 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
/* { dg-require-effective-target vect_int } */
/* { dg-require-effective-target arm_v8_2a_dotprod_neon_hw { target { aarch64*-*-* || arm*-*-* } } } */
/* { dg-additional-options "-march=armv8.2-a+dotprod" { target { aarch64*-*-* } } } */
/* { dg-add-options arm_v8_2a_dotprod_neon }  */

#include <stdarg.h>
#include "tree-vect.h"

#define N 64

#define DOT1 43680

signed char X[N] __attribute__ ((__aligned__(__BIGGEST_ALIGNMENT__)));
signed char Y[N] __attribute__ ((__aligned__(__BIGGEST_ALIGNMENT__)));

/* char->short->int dot product.
   The dot-product pattern should be detected.
   Vectorizable on vect_sdot_qi targets (targets that support dot-product of 
   signed chars).

   In the future could also be vectorized as widening-mult + widening-summation,
   or with type-conversion support.
 */
__attribute__ ((noinline)) int
foo1(int len) {
  int i;
  int result = 0;
  short prod;

  for (i=0; i<len; i++) {
    prod = X[i] * Y[i];
    result += prod;
  }
  return result;
}

int main (void)
{
  int i, dot1;

  check_vect ();

  for (i=0; i<N; i++) {
    X[i] = i;
    Y[i] = 64-i;
    __asm__ volatile ("");
  }

  dot1 = foo1 (N);
  if (dot1 != DOT1)
    abort ();

  return 0;
}

/* { dg-final { scan-tree-dump-times "vect_recog_dot_prod_pattern: detected" 1 "vect" } } */
/* { dg-final { scan-tree-dump-times "vect_recog_widen_mult_pattern: detected" 1 "vect" } } */
/* { dg-final { scan-tree-dump-times "vectorized 1 loops" 1 "vect" { target vect_sdot_qi } } } */
/* { dg-final { scan-tree-dump-times "vectorized 1 loops" 1 "vect" { target { vect_widen_mult_qi_to_hi && vect_widen_sum_hi_to_si } } } } */